Hum. Reprod. Advance Access originally published online on October 6, 2005
Human Reproduction 2006 21(1):68-79; doi:10.1093/humrep/dei313
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Developmental and reproductive performance in circadian mutant mice
1 Department of Anatomy, Downing Street, Cambridge CB2 3DY, UK and 2 MRC Laboratory of Molecular Biology, Hills Road, Cambridge CB2 2QH, UK
3 To whom correspondence should be addressed. E-mail: mhj{at}mole.bio.cam.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
BACKGROUND: Genes underlying circadian rhythm generation are expressed in many tissues. We explore a role for circadian rhythms in the timing and efficacy of mouse reproduction and development using a genetic approach. METHODS: We compare fecundity in ClockD19 mutant mice (a dominant-negative protein essential for circadian rhythm activity) and in Vipr2–/– null mutant mice (affecting the generation and output of the circadian rhythm of the hypothalamic suprachiasmatic nucleus) with wild type (WT) litter mates under both a 12 h:12 h light:dark cycle and continuous darkness. RESULTS: Uteri from ClockD19 mice show no circadian rhythm and Vipr2–/– mice show a phase-advanced rhythm compared to WT uteri. In neither mutant line were homozygous or heterozygous fetuses lethal. Sexually mature adults of both mutant lines showed mildly reduced male in vivo (but not in vitro) fertility and irregular estrous cycles exacerbated by continuous darkness. However, pregnancy rates and neonatal litter sizes were not affected. The ClockD19 mutant line was distinguishable from the Vipr2–/– null mutant line in showing more peri-natal delivery problems and very poor survival of offspring to weaning. CONCLUSIONS: A fully functional central and peripheral circadian clock is not essential for reproduction and development to term, but has critical roles peri-natally and post-partum.
Key words: circadian rhythm/clock mutants/fecundity/mice/uterus
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Circadian rhythms play a fundamental role in the effective functioning of complex organisms by allowing the body to anticipate changing environments in ways that enhance survival (Pittendrigh, 1993
). Conversely, defective rhythmicity places organisms at a disadvantage, e.g. the debilitating effects of jet-lag, shift working and genetic sleep disorders on behaviour and physiology (Hastings et al., 2003). In recent years, a basic description of the molecular clock that drives the endogenous circadian rhythm has been achieved. It consists of oscillatory feedback activities of a small group of circadian canonical clock-work gene products (King and Takahashi, 2000). Additionally, a large number of downstream genes has been shown to be controlled directly or indirectly by the oscillatory activity of these canonical genes, thereby imparting rhythmicity to tissue activities (Akhtar et al., 2002; Panda et al., 2002; Storch et al., 2002). Although most work initially assumed that the hypothalamic suprachiasmatic nucleus (SCN) was the locus of the endogenous clock, it has now become clear that the same or very similar clockwork molecular mechanisms function in most cells and tissues of the body (Yamazaki et al., 2000; Akhtar et al., 2002), and that these clocks can continue to run in the absence of an SCN (Yoo et al., 2004). The role of the SCN is now seen as coordinating the body’s multiple clocks through a variety of neural and humoral mechanisms (Guo et al., 2005) to keep them in an appropriate phase relationship to one another, such that whole body function is effectively integrated (Nagoshi et al., 2004). The SCN clock’s role is seen as important because it provides the major route by which external environmental circadian changes, principally the light/dark cycle, can be used to gain synchronization of the phase and period of the endogenous body clocks with solar and calendrical time. However, this hierarchical model may underestimate the capacity of the peripheral clocks to respond to environmental inputs independent of SCN influence (Carr et al., 2003).
Timing and rhythms are important during reproduction and development (Johnson and Day, 2000; Day et al., 2001; Johnson, 2001). The circadian rhythm and the SCN are intimately involved in timing of the LH surge during the estrous cycle (Alleva et al., 1971; Turek et al., 1984; Palm et al., 2001b; Barbacka-Surowiak et al., 2003; Gerhold et al., 2005), the initiation of fecundity at puberty (Kriegsfeld et al., 2002), and also, via the pineal, influence seasonal changes in fertility (Nuesslein-Hildesheim et al., 2000; Lincoln et al., 2003). The coordination of maternal circadian rhythmicity with that of fetus and neonate may be important for parturition and early post-natal survival (Kennaway, 2002). The canonical circadian genes are expressed in the uterus (Johnson et al., 2002), oviduct (Kennaway et al., 2003), testis (Alvarez et al., 2003; Bittman et al., 2003; Morse et al., 2003) and preimplantation embryo (Johnson et al., 2002; Hamatani et al., 2004). Their expression in both male and female tract and in the embryo has led to the conjecture that they might play a role in mammalian reproduction, particularly in the timing (or mis-timing when disordered) of early pregnancy and implantation (Johnson, 2001). A role in reproduction has been proposed in Drosophila, in which mutations in canonical circadian genes have been shown to impair reproductive fitness (Beaver et al., 2002, 2003; Beaver and Giebultowicz, 2004). There is indeed anecdotal evidence that mutations/deletions of clock-work genes are associated with poor reproductive performance (Herzog et al., 2000; Low-Zeddies and Takahashi, 2001; Chappell et al., 2003) and exposure of mice to short (22 h) or long (26 h) days leads to elevated embryonic loss (Endo and Wanatabe, 1989). A recent study of reproductive function also suggested impaired performance (Miller et al., 2004). In this paper, we go further and describe the impact of two distinctive genetic disturbances of circadian timing on various aspects of male and female fecundity. We also compare fecundity in normal light:dark (L:D) conditions with that under conditions in which external lighting rhythms were absent (continuous darkness or D:D) to reveal the full impact of circadian incompetence. The two mutant lines that differ in the locus of their circadian defect are Clock
19 and Vipr2–/–.
The Clock19 mutant line was generated through in vivo ethylnitrosourea (ENU) mutagenesis, resulting in an A
T mutation in the splice donor site downstream of exon 19, which leads to exon skipping and the elimination of 51 amino acids in the C-terminal region of the CLOCK protein (Vitaterna et al., 1994; King et al., 1997a,b). This deletion removes the ability of the Brain and Muscle Amt-Like protein 1 (BMAL1):CLOCK19 dimer to initiate transcription at an E-box (Gekakis et al., 1998), and does so in both the SCN and peripheral tissues. Thus, in the Clock19 mutant line, rhythms in both tissues are compromised, as evidenced by their loss of the oscillatory expression of the Per, Cry and Bmal1 genes at both sites (Kume et al., 1999; Oishi et al., 2000; Minami et al., 2002), with knock-on effects on many downstream circadian-controlled genes (Panda et al., 2002) and on circadian behaviour (Easton et al., 2000). In contrast, the VPAC2 receptor is not a canonical circadian gene, and thus not involved in the primary generation of circadian rhythms in the SCN or peripheral tissues. It does, however, function as a receptor for vasoactive intestinal polypeptide (VIP) and adenylate cyclase-activating polypeptide (PACAP; Harmar et al., 1998), both of which are implicated in the photic entrainment of circadian rhythms in the SCN (Cutler et al., 2003) and in mediation of GnRH secretion (Gerhold et al., 2005). In mice with a null mutation of the VPAC2 receptor gene (Vipr2–/–), the circadian rhythm within the SCN is grossly deficient (Harmar et al., 2002), leading to defective SCN clock-work. In contrast, the peripheral tissue molecular clock-work should not be directly affected but only indirectly via influences from the SCN or elsewhere. Thus, comparison between these mutants may help distinguish any central SCN effect on fecundity from a combined central and peripheral defect.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Animal care and handling
All work was conducted and licensed under the Animals (Scientific Procedures) Act, 1986 and had local ethical approval. Mutant lines: Clock
19 mice were obtained from the Jackson Laboratory (Maine, USA; King et al., 1997a,b; http://www.informatics.jax.org/javawi2//servlet/WIFetch?page=searchTool&query=clock&selectedQuery=Genes+and+Markers) and Vipr2–/– mice from Prof. A.J.Harmar (Department of Neuroscience, University of Edinburgh, 1 George Square, Edinburgh EH8 9JZ, UK; Harmar et al., 2002). Both lines are on C57Bl6 backgrounds and had been maintained since formation by in-breeding through heterozygous crosses. Lines were maintained in a 12:12 h light/dark cycle, the dark period illuminated by dim red light (<20 lux). Food and water were available ad libitum. Offspring were chipped (IdentiChip, Dunnington, York, UK) and tailed for identification and genotyping respectively.
PCR genotyping
Tail tip DNA was extracted and 20 ng used for PCR analysis. For the Clock19 line, the primer sets used (King et al., 1997a) were 5'-GGTCAAGGGCTACAGGTA-3' plus 5'-TGGGGTAAAAAGACCTCTTGCC-3' anticipating a product size of 153 bp from wild type (WT) mouse DNA, and 5'-AGCACCTTCCTTTGCAGTTCG-3' plus 5'-TGTGCTCAGACAGAATAAGTA-3' anticipating a product size of 355 bp from Clock19 DNA samples (Sigma–Genosys, Cambridge, UK). The 25 µl PCR reaction mixture contained 2 mmol/l MgCl2, 0.2 mmol/l dNTP, 0.02 µmol/l of primers, 20 ng of DNA sample, 0.1 IU of Bioline Taq and 2.5 µl of 10 x Bioline buffer (Bioline, London, UK). The PCR cycle was 94°C for 1 min and followed with 30 cycles at 94°C for 30 s, 53°C for 30 s and 72°C for 1 min and then end at 72°C for 7 min. The PCR product was then run on 2% agarose gel and stained with ethidium bromide. For the Vipr2–/– line, two primer sets were used courtesy of Prof A.J.Harmar: 5'-TCAGAGGGAAGTAGGGGTGGAAGGAGGGACG-3' plus 5'-TACCTCTCTGATTCTCCGTTTGGCTGCTTAGC-3' with anticipated product sizes of 2.5 kbp from WT DNA and 7.5 kbp from homozygous Vipr2–/– mutant DNA, and 5'-CGCTTCCTCGTGCTTTACGGTATCGCCGCTCC-3' plus 5'-TCCCCACTGTCACAAGGCTACATTAGTTTTGC-3' with no expected product from WT DNA and a 2.5 kbp product from Vipr2–/– homozygous mutant DNA samples. The 25 µl PCR reaction mixture contained 21.25 nmol/l MgCl2, 15 mmol/l dNTP, 25 µmol/l of primers, 50 ng of DNA sample, 1.875 IU of Taq (Roche Expand long template PCR system, Hertfordshire, UK) and 2.5µl of 10 x Roche buffer 3. The PCR cycle was 94°C for 2 min and 10 cycles of 94°C for 10 s, 65°C for 30 s and 68°C for 12 min and then 20 cycles of 94°C for 10 s, 65°C for 30 s and 68°C for 12 min with 20 s extension and was then ended with 7 min at 68°C. The PCR product was then run on 0.8% agarose gel and stained with ethidium bromide.
Estrus and activity assessment
Females were housed in single cages fitted with activity wheels. Rotation of the wheel activated a switch, which was recorded and binned in 6 min blocks (Viglen HD40V; Viglen Computers, London, UK) using software Dataquest IV software (Minimitter Co., Sun River, Oregon, USA). Light control (12 h light:12 h dark) was achieved by placing cages into a purpose-built, ventilated, light-sealed box. The photo-schedule involved a normal L:D cycle and release into constant darkness to free run for 
20 days. Estrous cycle determination was by vaginal flushing daily during the dark period to avoid disturbing the activity of the mouse during its inactive, light phase.
Assessment of male reproductive capability
Sperm count
We adapted the method of Baisong and Bishop (2003). Both caudal epididymides were removed and each was placed in 0.5 ml of Whittingham’s medium + 30 mg/ml BSA (W+30; made in the laboratory) that had been incubated overnight at 37°C in 5% CO2 under mineral oil. Using two pairs of fine forceps, each cauda epididymis was divided into four pieces and incubated in 5% CO2 incubator at 37°C for 20 min for dispersion of the sperm. Five microlitres of the dispersed sperm were removed, fixed in 45 µl of 4% paraformaldehyde and mixed by pipetting. Sperm counts were made using a haemocytometer and are expressed as sperm number per ml of medium.
IVF
Sperm from the vasa deferentia were extruded into 0.25 ml pre-equilibrated W+30 under oil and incubated at 37°C in 5% CO2 for 2 h to capacitate. Oocytes were collected from MF1 out-bred females (Harlan) superovulated by 10 IU PMS (Genus Express, Bury St Edmunds, UK) followed after 48 h by 10 IU HCG (Genus Express), 14–15 h post-hCG into 0.5 ml of pre-equilibrated W+30 under oil. Fifty microlitres of capacitated sperm were added and after 4 h the oocytes were placed in a M16+BSA medium under oil in 37°C incubator at 5% CO2 for 24 h, before scoring for fertilization as assessed by cleavage to 2-cells.
Sperm motility
Released vas sperm were scored subjectively, immediately after recovery and after 2 and 4 h in vitro, as +++ when most were very active, as ++ when the most were mildly active or less than half were strongly active, and as + when few sperm were active or most were poorly active. No motility was scored as –.
Coagulation test
Contents of the coagulating gland and seminal vesicle were mixed on a slide and coagulation scored visually as a measure of the capacity to form a copulatory plug.
Male fertility testing
Male fertility was tested by pairing a single homozygous Clock19 or Vipr2–/– male (or WT male litter mate) with an out-bred MF1 female (5–6 weeks old; Harlan, UK). Cages were inspected daily and the dates of littering of each female noted. Pups were counted, weighed, sexed and removed. The average litter number, size and inter-litter interval were calculated for each pair and the means compared using an unpaired t-test.
Fertility testing of mutant lines
Pairings were set up to allow closely scrutinized comparison of fertility of WT x WT crosses, mutant x mutant crosses, and WT males by mutant female crosses. These combinations provided for mutant females with offspring either wholly mutant (no WT genes during pregnancy) or with WT genes present in offspring but absent from the mother. All pairings were examined under both L:D and D:D conditions, since circadian mutant phenotypes are often best revealed under free-running conditions. Females were weighed daily to monitor pregnancy, inspected for vaginal plugs for evidence of mating, and inspected daily for litter birth and maintenance. Pairs in which no pregnancies/young at all were produced were eliminated from consideration for data analysis purposes (Clock line: 1x WT male + Clock19/Clock19 female in L:D and 2 in D:D; Vipr2 line: 1 x WT+WT in L:D; and 2 x WT male + Vipr2–/–/Vipr2–/– female in D:D). Litters were weaned at 3 weeks and weighed. Analysis of the reproductive time-course data for each pair categorized each day as: non-pregnant and pregnant (subdivided into pregnant to delivery and lost pregnancy). Where a female was suckling young, the first 4 days post-partum were not included in the non-pregnant data set to allow for lactational delay effects. In addition, data from our breeding colonies, which were not so closely monitored but provide larger numbers, were extracted for comparison with data from the closely monitored pairs.
Quantitative real-time PCR
Estrous mice maintained in 12 h L:12 h D were killed at 4 h intervals from ZT2 to ZT22 and uteri collected. Total RNA was extracted by crushing mouse uteri under liquid nitrogen using mortar and pestle. Frozen samples were then mixed with 500 µl Trizol (InVitrogen, Paisley, UK) and allowed to stand for 5 min. The mixture was centrifuged for 10 min at maximum speed, the supernatant separated and mixed with 100 µl of chloroform. After vigorous hand-mixing and centrifuging at 12 000 rpm (10,000 g) for 2 min, the supernatant was recovered, 12 µg linear acrylamide (Ambion cat. No. 9520, Huntingdon, UK) added, and RNA extracted using the RNeasy Mini kit (Qiagen, Crawley, UK).
Total RNA (800 ng) was reverse-transcribed using Moloney murine leukaemia virus (M-MLV) reverse transcriptase (Promega, Southampton UK) and pdN6 primer (InVitrogen, Paisley, UK). Primer/probes for real-time PCR were designed using Beacon Designer 3.1 software (Premier Biosoft Int., CA, USA) and synthesized by Sigma–Genosys (Cambridge, UK). Primers/probe were as follows: Bmal1 (GenBank accession no. AB012601 [GenBank] ) forward 5'-CCCACAGCATGGACAGCAT-3', reverse 5'-CTGGAATGCCTGGGACAGTG-3', probe 5'-FAM-CTGCCCTCTGGAGAAGGTGGCCA-TAMRA-3', product 80 bp; Per2 (GenBank accession no. AF035830 [GenBank] ) forward 5'-TCCCACCAGTCCCACCAAG-3', reverse 5'-TTCCCATTGTCGTCGCAGTC-3', probe 5'-FAM-CTGCC-CTGAGAGTCCCGTCCCGTG-TAMRA-3', product 149 bp; Cry1 (GenBank accession no. AB000777 [GenBank] ) forward 5'-AAGGAACGAGATGCAGCTATCAA-3', reverse 5'-GATCTTGTCCAGGTCATACAGTGT-3', probe 5'-FAM-CGCACGATGACTTCCACGCCAGCC-TAMRA-3', product 96 bp.
Per2 and Cry1 primers/probe do not cross the exon–exon boundary. The RNA samples were hence taken through the reverse transcription in the absence of M-MLV (–M-MLV) to allow assessment of any DNA contamination in the real time RT–PCR amplification step. The values obtained from –M-MLV were in all cases <5% of the reverse-transcribed samples and were subtracted from the cDNA data of the same sample to remove minor DNA contamination influence. The Bmal1 primers/probe cross an exon–exon boundary avoiding possible DNA contamination problems.
The real-time PCR amplification was carried out in a 36 µl reaction mixture containing 36 pmol of primers, 18 pmol of probe and 1 x Abgene QPCR buffer mix (ABgene, Surrey, UK). The real-time quantitative RT–PCR was carried out using a DNA engine Opticon 2 thermal cycler (MJ Research, GRI, Braintree, UK). Each sample assay was triplicated technically, and involved an initial phase at 50°C for 2 min and 95°C for 15 min, followed by 40 cycles of 15 s at 95°C and 1 min at 60°C. Fluoresence was detected using Opticon Monitor 2.02 (MJ Research).
Data were collected from four biological replicates for each time-point in both WT and Clock19 mutant lines and three biological replicates for the Vipr2–/– samples. Individual sample sets with all time-point samples were run together and each sample was proportioned to the mean value for all samples in that biological replicate to allow for absolute differences in measures for the different replicate runs. The results were normalized to the expression of 18S rRNA in the same samples diluted 1:100 to allow for the relatively high abundance of 18S using pre-developed assay reagents (Applied Biosystems, Warrington, UK). The data were analysed for significant circadian excursions by one-way analysis of variance.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Analysis of circadian molecular clock-work expression profiles in the uterus
The temporal expression pattern of Per2, Cry1 and Bmal1 mRNA in uteri from WT and homozygous mutant (Clock
19 and Vipr2–/–) mice is shown in Figure 1. The expected anti-phase profile of Cry/Per with Bmal is seen in the WT animals (Figure 1a), establishing for the first time that there is indeed a peripheral rhythm in this organ. In the mutant Clock19 females (Figure 1b), the rhythm is lost, as would be predicted from the presence of a dominant negative Clock gene product in both SCN and uterus. In contrast, the Vipr2–/– mutant uteri (Figure 1c) retained a clear rhythm but with a phase advance of 4–6 h in which all three gene products were affected as a cohort. Thus, real differences in the uterine rhythms (suppression for Clock19 and phase advanced for Vipr2–/–) are observed for each line in relation to WT controls, validating our experimental approach.
|
Viability of fetuses of differing genetic make-up
In order to determine whether the genotype of the embryo/fetus significantly affects viability to birth, the breeding data from heterozygote crosses for each line were analysed. The results in Table I show that there is no clear deficit of heterozygous or homozygous mutant offspring in either line, suggesting that no significant prenatal lethality associated with fetal genotype is occurring.
|
Male fertility
Homozygous Clock19 (n = 10) or Vipr2–/– (n = 5) males (or control +/+ male litter mates; n = 10 and 5 respectively) were paired individually with a single out-bred MF1 female (Clock19 for 272 days and Vipr2–/– for 136 days). Two Clock19 males produced no litters and were excluded from further analysis. Circadian mutation had a significant impact on male fecundity in so far as mean litter size in both lines was significantly lower for homozygous mutant males than for their WT litter mates (Table II). There was, however, no significant difference for either line in the mean litter number and inter-litter interval between homozygous mutant and control WT males. This result suggests that mating frequency was unaffected. A sperm origin for the differences in litter size was not, however, evident from in vitro tests. No significant difference in sperm count, sperm fertilizing capacity in vitro and testis weight was observed between the Clock19 and Vipr2–/– mutants and their respective WT controls (Table II). There was also no difference between the homozygous mutants and their control WT in sperm motility and coagulation properties (data not shown). Thus, we were unable to locate the site of the mild reduction in male fecundity.
|
The non-pregnant female reproductive cycle
Homozygous Clock19 (n = 13) and Vipr2–/– (n = 7) females, and control WT female litter mates (n = 10 and 7 respectively) were housed in activity cages under L:D or D:D conditions, and vaginally flushed daily. Representative actograms (with periods of estrus marked) are shown in Figure 2 for a homozygous Clock19 female and a WT control in both L:D and D:D conditions and confirmed that circadian activity patterns are lost only in continuous darkness. Note the absence of any evidence for an estrogen-dependent phase advance of the activity on the day of estrus, in contrast to previous reports for the hamster and rat (Alleva et al., 1971; Wollnik and Turek, 1988; Morin et al., 2002). Estrous cycles from individual females are shown in Figure 3, and reveal that females of both mutant lines in L:D showed a significantly lengthened mean cycle length compared with control WT females (Table III). In D:D the estrous cycle of both mutants was significantly longer than 4 days compared with their WT control under both lighting conditions. In the Clock19 females under both lighting conditions, the main contributor to the increased cycle length was the period of estrus (Table III and Figure 3). Vipr2–/– females had significantly extended estrus under D:D conditions, although this was offset to some extent by a reduced pro-estrus. Thus, the estrous cycle is destabilized by both mutations, and the effect is very pronounced in D:D, suggesting that a L:D lighting cycle can to some degree compensate for the mutational effects, but less so than the activity compensation provided by L:D conditions (Figure 2).
|
). Note that estrus is of variable length in mutant females regardless of lighting conditions. Also note absence of evidence of phase advance in activity in WT female on day of estrus.
|
|
Comparative fertility
Having analysed the reproductive characteristics of males and females separately, the fertility of both lines under both L:D and D:D conditions was compared in three mating combinations: inter-crossed WT, inter-crossed homozygous mutant, and WT male x homozygous mutant female. Despite the irregularity of their cycles, Clock19 mutant females did not remain free of pregnancy substantially longer than WT females under either lighting condition (Table IV). In contrast, Vipr2–/– females under L:D conditions were non-pregnant proportionately longer than were WT controls. Constant darkness increased non-pregnant time significantly for both lines, but mutant and WT controls were affected similarly.
|
Where pregnancy did occur, some pregnancy losses pre-term were observed in both lines under L:D conditions, but these losses did not correlate significantly with maternal genotype (mutant versus WT) nor mating combination (Table IV). However, under constant darkness, Clock19 x Clock19 pairings showed significantly higher pregnancy losses.
In both lines, mutant females had problems with term delivery (Table IV), and this was significantly different from WT for the Clock19 females (8/46; 17.4%), but not for the Vipr2–/– females (4/47; 8.5%). Typically, the problem took the form of a prolonged labour, which was monitored regularly and in one case led to termination of the breeding experiment for animal welfare reasons.
The mean numbers of pups born per litter and the numbers surviving to weaning for both Clock19 and Vipr2–/– mutant lines are shown in Figure 4. Under L:D conditions, no significant differences in numbers of pups born/litter among the different mating combinations are evident. However, for the Clock19 homozygous females (but not for Vipr2–/– females) paired with WT or mutant males, significantly fewer pups/litter were born under D:D conditions compared with L:D conditions. Clock19 females also showed a significant reduction in the survival of pups to weaning under both lighting conditions and regardless of whether the offspring carried a WT gene. In contrast, pup loss to weaning by the Vipr2–/– females was minimal and not different from controls.
|
The data from these daily monitored pairs were then compared with a more extensive data bank from our breeding colony of each line, both maintained under 12 h L:D cycles. The data are summarized in Table V and broadly support the more detailed analysis. Thus, for Vipr2–/– there are no significant differences in litter size between mutant pairs and WT control pairs, but for Clock19 with the larger numbers of breeding pairs analysed there is a small but significant difference between mutant and WT female. However, it is difficult given the less close monitoring to know whether the difference is due to undetected early post-natal loss rather than reduced litter size per se. In contrast, there is clear evidence for the selective post-natal loss for Clock19 x Clock19 pairs only.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
We have compared the fertility of mice with mutations that affect directly either the central hypothalamic circadian clock output alone (Vipr2–/–) or both central output and peripheral clock function (Clock
19). Comparisons were made with WT litter mates (to control for inbreeding effects on fertility) and under both light:dark and continuous darkness (to compare fertility with and without the main environmental circadian synchronizer).
Males from the two mutant lines were not distinguishable from one another, fertility being mildly reduced in both compared to WT, as expressed through modestly reduced litter sizes. Homozygous mutant males synthesize sperm (confirming for Clock19 males the study by Morse et al., 2003), which did not show evidence of subfertility in vitro, although any small deficit in sperm fertilizing capacity in vivo may have been masked by the relatively large numbers of sperm used for in vitro assays. The similar litter numbers and inter-litter intervals for mutant and WT males suggest no problems with male behaviour, mating frequency or pregnancy initiations and maintenance, so the decrease in pups per litter could be due to either a lower in-vivo fertilization (IVF) rate or a greater loss of heterozygous embryos. The latter is unlikely, given that the breeding data from the two lines do not provide any evidence for selective loss of heterozygous embryos compared to WT embryos, so a mild sperm fertility defect may be responsible. Previous studies have suggested that Per1 & 2, Bmal1, Clock and Cry1 are expressed in selected cells of the spermatogenic lineage and more controversially in Leydig cells, but their expression is constitutive and temporally invariant (Alvarez et al., 2003; Bittman et al., 2003; Morse et al., 2003), although one anomalous study did indicate possible circadian rhythmicity (Zylka et al., 1998). It is possible that some minor defect in sperm function or number does occur in the absence of functional CLOCK protein, as has been reported for fruit flies and moths (Giebultowicz and Riemann, 1990; Beaver et al., 2002). In moths, the fecundity defect has been related to a locally controlled circadian release of sperm from testis to vas deferens, but such studies in mice are lacking, so possibly a transport defect is also responsible in mice? The VPAC2 receptor is also expressed in the testis (mainly in the tunica albuginea), epididymis and vas deferens in association with smooth muscle (Harmar et al., 2004), where it is not involved in circadian rhythmicity. In its absence, young male mice are fertile, but older males are subfertile and all ages show metabolic disturbances (Asnicar et al., 2002), which may account for the slightly reduced fertility observed in this study. In conclusion, there is a clear similarity of reproductive performance deficit in both mutant lines, but we cannot conclude that it is caused in either line by the specific loss of testicular circadian clock-work activity. Non-specific deficits on general health may have a marginal fertility impact.
Female estrous cycles are clearly extended in Clock19 females due to a variable increase in the period spent in estrus. Effects in the Vipr2–/– females were less marked in L:D, but in both lines exposure of mutant (but not wild-type) females to continuous darkness sharply increased the variability of cycle length and significantly increased the period spent in estrus for Vipr2–/– females. Thus, in the absence of an environmental synchronizer, the effect of both circadian mutations on reproductive function is magnified. Ovulation in WT mice involves the interaction between high estrogen levels sensitizing the gonadotrophs (thereby specifying the day of LH release and ovulation) with a circadian signal from the SCN mediated via VIP innervation of the gonadotrophs (thereby specifying the time of day of ovulation; Krajnak et al., 1998; Kalsbeek and Buijis, 2002; Chappell et al., 2003; Gerhold et al., 2005). In both mutant lines, the circadian component of the input to the gonadotrophs will be compromised, but clearly the sustained high estrogen levels underlying the prolonged estrus cannot be. One would thus predict an absent or reduced LH surge in mutant females, and this has been reported for Clock19 females (Miller et al., 2004). How then is ovulation achieved when the females are paired with males? Presumably, the missing SCN input is replaced by some other input. Is perhaps a coital response to mating being revealed in mice in the absence of an SCN signal, as, for example, can be seen naturally to induce an LH surge in the estrous rabbit (Johnson and Everitt, 2000)? Or could leptin be elevated in the presence of a male to stimulate ovulation (Barkan et al., 2005)?
Thus, the studies on males and non-pregnant females do reveal impaired but not inhibited reproductive function in both sexes of both mutant lines. However, these defects do not prevent fertile matings between homozygous mutant mice. Pregnancy losses were observed in both lines, but under L:D conditions these were not obviously related to the genotype of the mother (WT or homozygous mutant) or of the fetuses (homozygous mutant, heterozygous mutant or WT), suggesting that the problem may be more related to a genetic background effect, possibly resulting from the inbreeding of the lines. Indeed, background effects may explain the somewhat contradictory published literature on pregnancy losses in the Clock19 line, with reports of no losses (Low-Zeddies and Takahashi, 2001; Kennaway et al., 2005) and fetal resorption rates of up to 75% by days 11–14 in mutant females compared to 30% by day 19 in WT controls (Miller et al., 2004). In our study, a genetically based difference in fetal loss emerged only under continuous darkness, homozygous mutant females of both lines carrying homozygous mutant pups showing increased pregnancy loss over other D:D pairings, although this was only significant for the Clock19 line. This is the only result that might suggest a function for an endogenous clock during pregnancy, in this case in fetuses helping to sustain a pregnancy in the absence of an exogenous L:D driver. However, we would not wish to press this point in light of the likely effects of genetic background described above, which typically can be variable in impact, and the fact that continuous darkness reduced the size of litters significantly for the mutant females in the Clock19 line (although not in the Vipr2–/– line) regardless of fetal genotype.
Delivery was problematic in both lines. There are two categories of explanation. Delivery could be impaired if some fetuses had died; in effect the delivery problem might represent a ‘late pregnancy loss’. Alternatively, there could be a real influence of the circadian clock-work on the timing of parturition. We incline to this interpretation because delivery was not problematic in the 53 WT female control term pregnancies. Delayed and/or problematic delivery in the Clock19 line has been reported previously (Miller et al., 2004; Kennaway et al., 2005), and evidence from SCN ablation studies in rodentine mothers and/or their fetuses has previously suggested the existence of a circadian gating of birth timing (Reppert et al., 1987). Studies suggest that the oxytocin sensitivity of the uterus may show circadian variation with an impact on the timing of parturition (Nathanielsz, 1998), and there is some evidence for circadian variations in human singleton spontaneous uncomplicated births (Mancuso et al., 2004). Taken together, these observations strongly suggest that the timing of birth is influenced by the circadian clock. Prostaglandin production is required for parturition and full fetal maturation in mice, and in its absence delayed and/or problematic delivery and post-natal losses occur (Reese et al., 2000), but an essential role for oxytocin is more controversial (Young et al., 1996; Gross et al., 2000; Douglas et al., 2002).
Thus, overall reproductive functions in mutant animals of the Clock19 and Vipr2–/– lines were very similar and differed from WT control animals in limited ways. The relative lack of effect on pregnancy in particular is surprising, given that the uterine expression patterns of circadian genes in both mutant lines was very different from that of controls. Thus, the temporal sequence of Per2, Cry1 and Bmal1 in WT uteri was comparable to that observed in other peripheral tissues such as liver and heart (Akhtar et al., 2002) (although unlike the atypical biphasic pattern of clock gene expression in the oviduct reported by Kennaway et al. 2003), and is likely to be reflected in the circadian oscillation of a large number of other downstream uterine genes (e.g. Kennaway et al., 2003; Horard et al., 2004). In contrast, no circadian rhythm was detected in Clock19 mice, consistent with the lack of a functional uterine CLOCK protein and a lack of molecular redundancy for this canonical gene. Interestingly, the rhythms of gene expression in uteri of Vipr2–/–, in which the Clock locus was WT and clearly able to function, were very well-defined and retained their internal phase relationships. Nevertheless, they were co-ordinately advanced relative to those of WT mice. A comparable phase advance of the circadian programme has also been observed recently in other peripheral tissues, including heart and liver, of Vipr2–/–mice (E.S.Maywood and M.H.Hastings, unpublished data). The presence of a uterine rhythm in the absence of a rhythmic outflow from the SCN seems likely to reflect indirectly the influence of other rhythmic factors to which these mice may be exposed, possibly related to feeding schedules (Damiola et al., 2000; Stokkan et al., 2001; Davidson et al., 2002)? Although feeding can cue circadian activity in the liver, it seems less likely perhaps to affect the uterus directly? Perhaps a secondary messenger conveys external periodicity indirectly and independently of the SCN? For example, prostaglandin E2 has been reported to have in vivo clock-resetting activity (Tsuchiya et al., 2005) and its receptors are present in the uterus (Yang et al., 1997), although whether and how feeding might programme PGE2 circadian rhythmicity is unclear.
However, what is most striking is that both of the genetic disturbances of the circadian rhythm in the uterus have so little effect physiologically. Thus, the only major difference between the lines did not obviously depend on the uterus. There was a very poor post-natal survival of both heterozygous and homozygous Clock19 (but not Vipr2–/–) pups born to homozygous mutant mothers. A much milder effect of the Clock19 mutation of post-natal survival has been reported previously (Kennaway et al., 2005). The pup loss occurred early and appeared to be related to inadequate feeding due to poor milk production. Failure to produce (prolactin; Naylor et al., 2005) or eject (oxytocin; Young et al., 1996) milk would explain these results. Links between the circadian system and maternal prolactin release have been reported (Palm et al., 2001a; Wall et al., 2005), and it is possible that defective temporal alignment of suckling stimuli and prolactin secretion leads to inadequate milk production. However, it is unclear why the Clock19 line should be more affected post-natally than the Vipr2–/– line, unless a circadian clock activity in the mammary tissue and/or the neonate is involved. If the latter were the explanation, then heterozygous offspring might be expected to do better than homozygous mutant offspring, but there was no evidence for this, suggesting a maternal (mammary tissue?) locus of defect.
In conclusion, fully functional central and peripheral circadian clocks appear to have a role in the timing of estrous/ovulation, of parturition and of lactation, but do not appear to be essential for initiation and maintenance of pregnancy to term despite being expressed rhythmically in the uterus. Perhaps these findings are not surprising in that each of the three affected reproductive features involves clear interactions between a mother and her environment. Once underway, pregnancy and embryo–fetal development are essentially a self-contained system. Whilst developmental timing is clearly important, any clock mediating it is not critically dependent on the use of circadian genes expressed rhythmically in either mother or embryo–fetus. Whether some of these genes have other non-circadian timing roles, as perhaps they do in the testis, is still an open question.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
We thank Tony Harmar for supplying the nuclear breeding stock of the Vipr2–/– line, and Adrian Woodhouse and Tracie Butcher and their staffs for caring for the mouse lines. The work was supported by a grant from the Wellcome Trust to M.H.J. and M.H.H. and a BBSRC studentship to H.D.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Akhtar RA, Reddy AB, Maywood ES, Clayton JD, King VM, Smith AG, Gant TW, Hastings MH and Kyriacou CP (2002) Circadian cycling of the mouse liver transcriptome, as revealed by cDNA microarray, is driven by the suprachiasmatic nucleus. Curr Biol 12,540–550.[CrossRef][ISI][Medline]
Alleva J, Waleski M and Alleva F (1971) A biological clock controlling the estrous cycle of the hamster. Endocrinology 88,1368–1379.[ISI][Medline]
Alvarez J, Chen D, Storer E and Sehgal A (2003) Non-cyclic and developmental stage-specific expression of circadian clock proteins during murine spermatogenesis. Biol Reprod 69,81–91.
Asnicar M, Koster A, Heiman M, Tinsley F, Smith D, Galbreath E, Fox N, Ma Y, Blum W and Hsiung H (2002) Vasoactive Intestinal Polypeptide/Pituitary Adenylate Cyclase-Activating Peptide Receptor 2 deficiency in mice results in growth retardation and increased basal metabolic rate. Endocrinology 143,3994–4006.
Baisong L and Bishop C (2003) Late onset of spermatogenesis and gain of fertility in POG-deficient mice indicate that POG is not necessary for the proliferation of spermatogonia. Biol Reprod 69,161–168.
Barbacka-Surowiak G, Surowiak J and Stoklosowa S (2003) The involvement of suprachiasmatic nuclei in the regulation of estrous cycles in rodents. Reprod Biol 3,99–129.[Medline]
Barkan D, Hurgin V, Dekel N, Amsterdam A and Rubinstein M (2005) Leptin induces ovulation in GnRH-deficient mice. FASEB J 19,133–145.
Beaver L and Giebultowicz J (2004) Regulation of copulation duration by period and timeless in Drosophila melanogaster. Curr Biol 14,1492–1497.[CrossRef][ISI][Medline]
Beaver L, Gvakharia B, Vollintine T, Hege D, Stanewsky R and Giebultowicz J (2002) Loss of circadian clock function decreases reproductive fitness in males of Drosophila melanogaster. Proc Nat Acad Sci USA 99,2134–2139.
Beaver L, Rush B, Gvakharia B and Giebultowicz J (2003) Noncircadian regulation and function of clock genes period and timeless in oogenesis of Drosophila melanogaster. J Biol Rhythms 18,463–472.[Abstract]
Bittman E, Doherty L, Huang L and Paroskie A (2003) Period gene expression in mouse endocrine tissues. Am J Physiol 285,R561–R569.
Carr A-J, Johnston J, Semikhodskii A, Nolan T, Cagampang F, Stirland J and Loudon A (2003) Photoperiod differentially regulates circadian oscillators in central and peripheral tissues of the Syrian hamster. Curr Biol 13,1543–1548.[CrossRef][ISI][Medline]
Chappell P, White R and Mellon P (2003) Circadian gene expression regulates pulsatile gonadotrophin-releasing hormone (GnRH) secretory patterns in the hypothalamic GnRH-secreting GT1-7 cell line. J Neurosci 23,11202–11213.
Cutler D, Haraura M, Reed H, Shen S, Sheward W, Morrison C, Marston H, Harmar A and Piggins H (2003) The mouse VPAC2 receptor confers suprachiasmatic nuclei cellular rhythmicity and responsiveness to vasoactive intestinal polypeptide in vitro. Eur J Neurosci 17,197–204.[CrossRef][ISI][Medline]
Damiola F, Minh NL, Preitner N, Kornmann B, Fleury-Olela F and Schibler U (2000) Restricted feeding uncouples circadian oscillators in peripheral tissues from the central pacemaker in the suprachiasmatic nucleus. Genes Dev 14,2950–2961.
Davidson A, Stokkan K-A, Yamazaki S and Menaker M (2002) Food-anticipatory activity and liver per1-luc activity in diabetic transgenic rats. Physiol Behav 76,21–26.[CrossRef][Medline]
Day ML, Winston N, McConnell JL, Cook DI and Johnson MH (2001) tiK+ toK+: an embryonic clock? Reprod Fertil Dev 13,69–79.[CrossRef][Medline]
Douglas A, Leng G and Russell J (2002) The importance of oxytocin mechanisms in the control of mouse parturition. Reproduction 123,543–552.[Abstract]
Easton A, Arbuzova J and Turek FW (2003) The circadian Clock mutation increases exploratory activity and escape-seeking behavior. Genes Brain Behav 2,11–19.[CrossRef][ISI][Medline]
Endo A and Wanatabe T (1989) Effects of non-24-hour days on reproductive efficiency and embryonic devleopment in mice. Gamete Res 22,435–441.[CrossRef][ISI][Medline]
Gekakis N, Staknis D, Nguen H, Davis F, Wilsbacher L, King D, Takahashi J and Weitz C (1998) Role of the CLOCK protein in the mammalian circadian mechanism. Science 280,1564–1569.
Gerhold L, Rosewell K and Wise P (2005) Suppression of vasoactive intestinal polypeptide in the suprachiasmatic nucleus leads to aging-like alterations in cAMP rhythms and activation of gonadotropin-releasing hormone neurons. J Neurosci 25,62–67.
Giebultowicz J and Riemann J (1990) Circadian system in the insect testes controls the rhythmic release of sperm. Prog Clin Biol Res 341B,655–660.
Gross G, Imamura T and Muglia L (2000) Gene knockout mice in the study of parturition. J Soc Gynecol Invest 7,88–95.[CrossRef][ISI][Medline]
Guo H, Brewer J, Champhekar A, Harris R and Bittman E (2005) Differential control of peripheral circadian rhythms by suprachiasmatic-dependent neural signals. Proc Natl Acad Sci USA 102,3111–3116.
Hamatani T, Carter M, Sharov A and Ko M (2004) Dynamics of global gene expression changes during mouse preimplantation development. Dev Cell 6,117–131.[CrossRef][ISI][Medline]
Harmar A, Arimura A, Gozes I, Journot L, Laburthe M, Plsegna J, Rawlings S, Robberecht P, Said S, Sreedharan S et al (1998) International Union of Pharmacology. XVIII. Nomenclature of receptors for vasoactive intestinal polypeptide and pituitary adenylate cyclase-activating polypeptide. Pharm Rev 50,265–270.
Harmar A, Marston H, Shen S, Spratt C, West K, Sheward W, Morrison C, Dorin J, Piggins H, Reubi J-C et al (2002) The VPAC2 receptor is essential for circadian function in the mouse suprachiasmatic nuclei. Cell 109,497–506.[CrossRef][ISI][Medline]
Harmar A, Sheward W, Morrison C, Waser B, Gugger M and Reubi J (2004) Distribution of the VPAC2 receptor in peripheral tissues of the mouse. Endocrinology 145,1203–1210.
Hastings M, Reddy A and Maywood E (2003) A clockwork web: circadian timing in brain and periphery, in health and disease. Nature Rev Neurosci 4,1–14.
Herzog E, Grace M, Harrer C, Williamson J, Shinohara K and Block G (2000) The role of Clock in the developmental expression of neuropeptides in the suprachiasmatic nucleus. J Comp Neurol 424,86–98.[CrossRef][ISI][Medline]
Horard B, Rayet B, Triqueneaux G, Laudet V, Delaunay F and Vanacker J-M (2004) Expression of the orphan nuclear receptor ERR is under circadian regulation in estrogen-responsive tissues. J Mol Endocrinol 33,87–97.[Abstract]
Johnson M (2001) Time and development. Reprod Biomed 4 (Suppl 1),39–45.
Johnson M and Day M (2000) Egg timers: how is developmental time measured in the early vertebrate embryo? BioEssays 22,57–63.[CrossRef][ISI][Medline]
Johnson M and Everitt B (2000) Essential Reproduction. 5th edn, Blackwell, Oxford, UK.
Johnson MH, Lim A, Fernando D and Day ML (2002) Circadian clockwork genes are expressed in the reproductive tract and conceptus of the early pregnant mouse. Reprod Biomed 4,140–145.
Kalsbeek A and Buijis R (2002) Output pathways of the mammalian suprachiasmatic nucleus: coding circadian time by transmitter selection and specific targetting. Cell Tissue Res 309,109–118.[CrossRef][ISI][Medline]
Kennaway D (2002) Programming of the fetal suprachiasmatic nucleus and subsequent adult rhythmicity. Trends Endocrinol Metab 13,398–402.[CrossRef][ISI][Medline]
Kennaway D, Varcoe T and Mau V (2003) Rhythmic expression of clock and clock-controlled genes in the rat oviduct. Mol Hum Reprod 9,503–507.
Kennaway D, Boden M and Voultsios A (2005) Reproductive performance in female Clock(Delta19) mutant mice. Reprod Fertil Dev 16,801–810.[CrossRef][Medline]
King D and Takahashi J (2000) Molecular genetics of circadian rhythms in mammals. Annu Rev Neurosci 23,713–742.[CrossRef][ISI][Medline]
King DP, Vitaterna MH, Chang AM, Dove WF, Pinto LH, Turek FW and Takahashi JS (1997a) The mouse Clock mutation behaves as an antimorph and maps within the W19H deletion, distal of Kit. Genetics 146,1049–1060.[Abstract]
King D, Zhao Y, Sangoram A, Wilsbacher L, Tanaka M, Antoch M, Steeves T, Vitaterna M, Kornhauser J, Lowrey P et al (1997b) Positional cloning of the mouse circadian Clock gene. Cell 89,641–653.[CrossRef][ISI][Medline]
Krajnak K, Kashon M, Rosewell K and Wise P (1998) Sex differences in the daily rhythm of vasoactive intestinal polypeptide but not arginine vasopressin messenger ribonucleic acid in the suprachiasmatic nuclei. Endocrinology 139,4189–4196.